betteib/xlm-tn-20epochs

This model is a fine-tuned version of Davlan/xlm-roberta-base-finetuned-arabic on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 6.7068
• Train Accuracy: 0.0290
• Validation Loss: 6.6661
• Validation Accuracy: 0.0290
• Epoch: 14

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 0.0001, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 0.0001, 'decay_steps': 18848, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 992, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.001}
• training_precision: float32

Training results

Train Loss	Train Accuracy	Validation Loss	Validation Accuracy	Epoch
9.7281	0.0031	9.4281	0.0045	0
9.2184	0.0051	8.9814	0.0060	1
8.7336	0.0068	8.4005	0.0084	2
8.1287	0.0109	7.7969	0.0133	3
7.6665	0.0159	7.4295	0.0222	4
7.3938	0.0233	7.1783	0.0289	5
7.2079	0.0287	7.0257	0.0286	6
7.0785	0.0292	6.9028	0.0291	7
6.9777	0.0294	6.8739	0.0287	8
6.9034	0.0292	6.8083	0.0281	9
6.8549	0.0292	6.8099	0.0280	10
6.7978	0.0292	6.7450	0.0286	11
6.7555	0.0291	6.7420	0.0281	12
6.7267	0.0293	6.6804	0.0291	13
6.7068	0.0290	6.6661	0.0290	14

Framework versions

• Transformers 4.31.0
• TensorFlow 2.12.0
• Datasets 2.19.1
• Tokenizers 0.13.3